Dr David Humphreys

dh2
Associate Professor of Evidence-Based Intervention and Policy Evaluation
Research Fellow at Green Templeton College
(+44) 01865 280339
david.humphreys@spi.ox.ac.uk

David Humphreys is an Associate Professor of Evidence-Based Intervention and Policy Evaluation. He is an interdisciplinary social scientist whose research spans several fields including:, criminology, social (or public) policy, public health, and epidemiology. His main topic of interest focuses on the causes, consequences and prevention of violence and injury. Much of his research investigates how structural changes—such as laws, regulations or changes to the built environment—impact on the rate and/or distribution of harm in the population.

David’s research typically utilises observational or non-randomised research designs. As a consequence, his research and teaching bring together an interest in methodological aspects of intervention research, such as: the design of complex interventions (natural and quasi-experiments); the measurement of physical harms and its related environmental exposures; and the assessment and synthesis of equity effects.

He joined the Department in October 2013, having previously held research positions at the Institute of Public Health and the Institute of Criminology at the University of Cambridge. He completed his doctorate at the Institute of Criminology, at the University of Cambridge, in 2011.

David has affiliations with the Violence Research Centre at the University of Cambridge and is affiliated with the Penn Injury Science Center at the Perelman School of Medicine at the University of Pennsylvania. He has held research grants from the NIHR, ESRC and the Joyce Foundation, among others.

David is on sabbatical in Hilary Term 2020

Research

David’s research interests include:

  • Violence and injury prevention (including: alcohol-related violence, gun violence, knife crime, intimate partner violence, child maltreatment, suicide, motor vehicle accidents, other)
  • Alcohol and drug misuse
  • Changes to the built environment and urban design
  • Physical inactivity
  • Methods for evaluating complex population level interventions (i.e. laws, regulations, policies)
Current project
Publications

  • Can synthetic controls improve causal inference in interrupted time series evaluations of public health interventions?

    2020
    |
    Journal article
    |
    Int J Epidemiol
    Interrupted time series designs are a valuable quasi-experimental approach for evaluating public health interventions. Interrupted time series extends a single group pre-post comparison by using multiple time points to control for underlying trends. But history bias-confounding by unexpected events occurring at the same time of the intervention-threatens the validity of this design and limits causal inference. Synthetic control methodology, a popular data-driven technique for deriving a control series from a pool of unexposed populations, is increasingly recommended. In this paper, we evaluate if and when synthetic controls can strengthen an interrupted time series design. First, we summarize the main observational study designs used in evaluative research, highlighting their respective uses, strengths, biases and design extensions for addressing these biases. Second, we outline when the use of synthetic controls can strengthen interrupted time series studies and when their combined use may be problematic. Third, we provide recommendations for using synthetic controls in interrupted time series and, using a real-world example, we illustrate the potential pitfalls of using a data-driven approach to identify a suitable control series. Finally, we emphasize the importance of theoretical approaches for informing study design and argue that synthetic control methods are not always well suited for generating a counterfactual that minimizes critical threats to interrupted time series studies. Advances in synthetic control methods bring new opportunities to conduct rigorous research in evaluating public health interventions. However, incorporating synthetic controls in interrupted time series studies may not always nullify important threats to validity nor improve causal inference.
    Interrupted time series, causal inference, history bias, quasi-experimental, synthetic controls

  • Assessing the impact of a local community subsidized rideshare program on road traffic injuries: an evaluation of the Evesham Saves Lives Program

    2020
    |
    Journal article
    |
    Injury Prevention

  • Evaluating the impact of penalising the use of mobile phones while driving on road traffic fatalities, serious injuries and mobile phone use: a systematic review.

    2020
    |
    Journal article
    |
    Inj Prev
    BACKGROUND: A vast literature has demonstrated that using mobile phones while driving increases the risk of road traffic crashes. In response, policy-makers have introduced bans and harsher penalties on using mobile phones while driving. Even though emerging evidence suggests that such measures may reduce mobile phone use and crashes, the literature has not been systematically reviewed and synthesised. OBJECTIVE: To evaluate the impact of penalising mobile phone use while driving on road traffic fatalities, serious injuries and the prevalence of mobile phone use while driving. METHODS: We employed a comprehensive search strategy using electronic databases, websites, handsearching and other sources to locate studies evaluating legislation on mobile phone use while driving. Randomised controlled trials, interrupted time series', controlled before-after studies with control(s) not exposed to harsher sanctions and panel data designs were included if they measured the outcomes of fatalities, serious injuries or the prevalence of mobile phone use while driving. Eligible studies were critically appraised. Due to substantial heterogeneity, the results were synthesised narratively. The synthesis structured studies according to the type of legislation and outcome measure. RESULTS: Of the 7420 studies retrieved, 32 were included. The evidence on the effects of penalising mobile phone use while driving was weak, and somewhat inconsistent, but pointed to a potential decrease in the prevalence of mobile phone use and fatalities for all-driver primary enforcement hand-held bans and texting bans. CONCLUSIONS: Preventing fatalities from risky driving practices may be helped by implementing harsher laws that penalise mobile phone use while driving.
    distraction, enforcement, legislation, motor vehicle � occupant, policy, systematic review

  • Universal background checks for handgun purchases can reduce homicide rates of African Americans

    2020
    |
    Journal article
    |
    Journal of Trauma and Acute Care Surgery

  • How do perceived and objective measures of neighbourhood disadvantage vary over time? Results from a prospective-longitudinal study in the UK with implications for longitudinal research on neighbourhood effects on health

    2020
    |
    Journal article
    |
    PLoS One

  • Ridesharing and motor vehicle crashes: a spatial ecological case-crossover study of trip-level data.

    2020
    |
    Journal article
    |
    Inj Prev
    BACKGROUND: Ridesharing services (eg, Uber, Lyft) have facilitated over 11 billion trips worldwide since operations began in 2010, but the impacts of ridesharing on motor vehicle injury crashes are largely unknown. - METHODS: This spatial ecological case-cross over used highly spatially and temporally resolved trip-level rideshare data and incident-level injury crash data for New York City (NYC) for 2017 and 2018. The space-time units of analysis were NYC taxi zone polygons partitioned into hours. For each taxi zone-hour we calculated counts of rideshare trip origins and rideshare trip destinations. Case units were taxi zone-hours in which any motor vehicle injury crash occurred, and matched control units were the same taxi zone from 1 week before (-168 hours) and 1 week after (+168 hours) the case unit. Conditional logistic regression models estimated the odds of observing a crash (separated into all injury crashes, motorist injury crashes, pedestrian injury crashes, cyclist injury crashes) relative to rideshare trip counts. Models controlled for taxi trips and other theoretically relevant covariates (eg, precipitation, holidays). RESULTS: Each additional 100 rideshare trips originating within a taxi zone-hour was associated with 4.6% increased odds of observing any injury crash compared with the control taxi zone-hours (OR=1.046; 95% CI 1.032 to 1.060). Associations were detected for motorist injury and pedestrian injury crashes, but not cyclist injury crashes. Findings were substantively similar for analyses conducted using trip destinations as the exposure of interest. CONCLUSIONS: Ridesharing contributes to increased injury burden due to motor vehicle crashes, particularly for motorist and pedestrian injury crashes at trip nodes.
    distraction, motor vehicle—non-traffic, motor vehicle—occupant, pedestrian

  • Increasing adolescent firearm homicides and racial disparities following Florida's 'Stand Your Ground' self-defence law.

    2020
    |
    Journal article
    |
    Injury prevention : journal of the International Society for Child and Adolescent Injury Prevention
    Establishing whether specific laws impact rates of firearm homicide in adolescents is critical for identifying opportunities to reduce preventable adolescent death. We evaluated Florida's Stand Your Ground law, enacted October 2005, using an interrupted time series design from 1999 to 2017. We used segmented quasi-Poisson regression to model underlying trends in quarterly rates of adolescent (15-19 years) firearm homicide in Florida and disaggregated by race (Black/White). We used synthetic and negative controls (firearm suicide) to address time-varying confounding. Before Florida's Stand Your Ground law, the mean quarterly rate was 1.53 firearm homicides per 100 000 adolescents. Black adolescents comprised 63.5% of all adolescent firearm homicides before and 71.8% after the law. After adjusting for trends, the law was associated with a 44.6% increase in adolescent firearm homicide. Our analysis indicates that Florida's Stand Your Ground is associated with a significant increase in firearm homicide and may also exacerbate racial disparities.

  • Long-term Exposure to Neighborhood Deprivation and Intimate Partner Violence Among Women: A UK Birth Cohort Study.

    2020
    |
    Journal article
    |
    Epidemiology (Cambridge, Mass.)
    BACKGROUND:Intimate partner violence is the most common form of violence perpetrated against women. To our knowledge, the effect of neighborhood disadvantage on intimate partner violence against women has never been investigated prospectively outside the United States. METHODS:We used data from the Avon Longitudinal Study of Parents and Children in the United Kingdom, which followed our target sample, 7,219 women, from birth and their mothers (from pregnancy). At age 21, 2,128 participants self-reported the frequency of experiencing physical, psychological, or sexual intimate partner violence since age 18. Participants' exposure to neighborhood-level deprivation and family-level socioeconomic characteristics (e.g., income) were measured at 10 time points from baseline (gestation) until children were 18 years old. We estimated the effect of cumulative exposure to greater neighborhood-level deprivation on the risk of experiencing intimate partner violence using marginal structural models with stabilized inverse probability weights, accounting for time-varying confounding by socioeconomic indicators and sample attrition. RESULTS:A one-unit increase in cumulative exposure to more severe neighborhood deprivation was associated with a 62% increase in participants' frequency of experiencing intimate partner violence (95% confidence interval 11%, 237%) and 36% increase in their risk of experiencing any intimate partner violence (95% confidence interval 1%, 85%). CONCLUSIONS:In our study, cumulative exposure to greater neighborhood deprivation over the first 18 years of life was associated with women's increased risk of experiencing intimate partner violence in early adulthood. Future studies should test this association across contexts, including underlying mechanisms, and evaluate preventive strategies that target structural disparities.

  • Long-term exposure to neighbourhood-level deprivation and the risk of experiencing intimate partner violence among women: a birth-cohort study in the United Kingdom

    2020
    |
    Journal article
    |
    Epidemiology

  • Child maltreatment and the risk of antisocial behaviour: A population-based cohort study spanning 50 years.

    2020
    |
    Journal article
    |
    Child abuse & neglect
    BACKGROUND:Child maltreatment is associated with an increased risk of antisocial behaviour; however, whether this risk persists and remains stable across the life-course is undetermined. OBJECTIVE:To examine associations between chid maltreatment and antisocial behaviour across the life-course. PARTICIPANTS AND SETTING:The study used 50 years of longitudinal data from the 1958 British birth cohort (n = 8088) measuring child neglect (prospectively) and abuse (retrospectively) and antisocial behaviour from childhood-to-adulthood. METHODS:Latent growth curve models analysed the longitudinal course of antisocial behaviour across childhood (7-16years) and adulthood (23-50years) as a function of child maltreatment. We used directed acyclic graphs to identify, and adjust for, potential confounders (biological, family, social). RESULTS:Child maltreatment was associated with higher levels of antisocial behaviour at all seven timepoints across the life-course (7-50years). Antisocial behaviour was elevated during childhood and adulthood in individuals who were maltreated, independently of confounding factors. Individuals who experienced multiple types of maltreatment were at the greatest risk of antisocial behaviour. Each additional maltreatment type was associated with an increased risk during both childhood (B = 0.173; SE = 0.024; p < .001) and adulthood (B = 0.137; SE = 0.014; p < .001). There was limited evidence that child maltreatment was associated with within-person rates of change, indicating that the increased risk of antisocial behaviour did not change over time. CONCLUSIONS:Child maltreatment is associated with an increased risk of antisocial behaviour, with a persistent and stable association remaining up to age 50. Our results highlight the burden of child maltreatment and the importance of providing long-term support for individuals who experience child maltreatment.
    • More

Further publications before 2013

Supervision

David would welcome applications for D.Phils from students who are committed to policy-relevant research on violence and injury prevention. He would be particularly interested to hear from students who are interested in examining the impact of structural level interventions, using quasi-experimental or other novel methodological designs.

Current students
Find out more

Read a summary of David's work on the Oxford Social Science Division site.